cs.CL（2024-10-20）

📊 共 13 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Causality for Large Language Models 综述：探讨因果关系融入大语言模型全生命周期的方法与未来方向 large language model foundation model
2 RoMemes: A multimodal meme corpus for the Romanian language 构建罗马尼亚语多模态表情包语料库RoMemes，促进AI理解网络迷因 multimodal
3 Ichigo: Mixed-Modal Early-Fusion Realtime Voice Assistant Ichigo：混合模态早期融合的实时语音助手，显著降低语音问答延迟。 large language model multimodal
4 When Machine Unlearning Meets Retrieval-Augmented Generation (RAG): Keep Secret or Forget Knowledge? 提出基于RAG的轻量级行为遗忘框架，解决LLM的敏感信息泄露问题 large language model multimodal
5 Hey GPT, Can You be More Racist? Analysis from Crowdsourced Attempts to Elicit Biased Content from Generative AI 通过众包提示工程分析生成式AI中的偏见诱导方法 large language model
6 Keep Guessing? When Considering Inference Scaling, Mind the Baselines 提出基于训练集答案频率的基线方法，评估LLM重复采样推理的真实提升 large language model
7 A Comprehensive Evaluation of Cognitive Biases in LLMs 大规模评估LLM认知偏差：构建通用测试框架与基准数据集 large language model ✅
8 LlamaLens: Specialized Multilingual LLM for Analyzing News and Social Media Content LlamaLens：面向新闻与社交媒体分析的专用多语言LLM large language model ✅
9 Redefining Proactivity for Information Seeking Dialogue 针对信息检索对话，提出基于新信息注入的主动性增强方法。 chain-of-thought
10 Lossless KV Cache Compression to 2% 提出CLLA架构，将KV缓存无损压缩至2%，显著提升LLM推理效率 large language model
11 "What is the value of {templates}?" Rethinking Document Information Extraction Datasets for LLMs 提出K2Q数据集，利用多样化模板提升LLM在文档信息抽取任务中的性能与鲁棒性 large language model
12 CalibraEval: Calibrating Prediction Distribution to Mitigate Selection Bias in LLMs-as-Judges CalibraEval：校准LLM预测分布以缓解LLM-as-Judges中的选择偏差 large language model
13 A Survey of Uncertainty Estimation in LLMs: Theory Meets Practice 综述LLM不确定性估计：理论与实践相结合，提升模型应用可信度 large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页